Method and System for Constructing Predictive Vehicle Driving Condition, Device and Medium Thereof

This application claims priority from the Chinese patent application 2024103297757 filed Mar. 22, 2024, the content of which is incorporated herein in the entirety by reference.

The present disclosure generally relates to the technical field of vehicles, in particular to a method and system for constructing a predictive vehicle driving condition, a device and a medium thereof.

A predictive vehicle driving condition refers to predicting a vehicle driving condition based on a position of an origin before departure. At present, methods for predicting the vehicle driving condition mainly include a method based on a Kalman filter model, a linear regression model, a Markov chain method and the like.

However, these traditional prediction methods merely consider one type of data, either historical driving data or real-time driving data. Moreover, the traditional prediction methods involve complex and cumbersome computations during the prediction process, which increases the likelihood of errors, resulting in low accuracy of final prediction results. Thus, a method and system for constructing a predictive vehicle driving condition, a device and a medium are provided to solve the above problems.

In view of the above defects or shortcomings in the prior art, the present disclosure provides a method and system for constructing a predictive vehicle driving condition, a device and a medium thereof, which can improve prediction accuracy and effectively support improvement in a vehicle driving range prediction algorithm and calibration optimization of a predictive vehicle energy management strategy.

In a first aspect, the present disclosure provides a method for constructing a predictive vehicle driving condition, including the following steps:

acquiring a first road set of a planned route from a map navigation system of a vehicle and retrieving historical vehicle driving data stored in the map navigation system, wherein the first road set at least includes a plurality of road segments identified by a global positioning system (GPS) and corresponding parameter data; and the historical vehicle driving data at least includes a plurality of point-in-time and corresponding driving data;

calculating first distances from longitudes and latitudes of the point-in-time to all roads in the first road set, measured by a satellite for the GPS to obtain a first distance set, and meanwhile, screening distances less than or equal to a distance threshold from the first distance set, to obtain a second distance set; generating a road-point-in-time set according to the second distance set identified by the GPS, wherein the road-point-in-time set at least includes a plurality of road segments and a plurality of point-in-time corresponding to each road segment;

sequencing a plurality of point-in-time corresponding to the same road segment in the road-point-in-time set according to a time sequence, to obtain a data sequence, meanwhile, traversing the data sequence, and cutting at two adjacent point-in-time with a time difference greater than a time threshold, to obtain a plurality of road segments corresponding to a current road;

establishing a road segment database of the same road segment according to the segments meeting a first preset condition in the road segment, wherein the road segment database at least includes a plurality of road segments and corresponding segment feature data;

screening segments with the segment feature data meeting a second preset condition from all the road segment databases, to generate a road-condition segment set, wherein the road-condition segment set at least includes a plurality of road segments and a plurality of condition segments corresponding to each road segment; and

sequencing a plurality of condition segments corresponding to each road segment in the road-condition segment set according to a sequence from a high confidence level to a low confidence level, to obtain a plurality of segment sequences, and meanwhile, screening condition segments meeting a third preset condition from all the segment sequences, to generate an optimal driving condition; and outputting the optimal driving condition to a display and applying the optimal driving condition to drive the vehicle.

According to the technical solution provided by the present disclosure, after obtaining the plurality of segment sequences and before screening the condition segments meeting the third preset condition from all the segment sequences, the method further includes the following steps:

sequencing the roads according to the planned route, to obtain a road sequence; and

screening the condition segments meeting the third preset condition from all the segment sequences according to the road sequence; wherein the third preset condition is defined as follows: a speed difference between the condition segments of two adjacent road segments is less than a preset speed difference.

According to the technical solution provided by the present disclosure, the screening the condition segments meeting the third preset condition from all the segment sequences according to the road sequence, specifically includes the following steps:

selecting a road corresponding to a position of an origin of the planned route from the road sequence, to serve as an initial traversal road;

acquiring a segment with a maximum chi-square value confidence level from the segment sequence corresponding to the initial traversal road, to serve as an initial segment;

traversing the segment sequence of the road connected with the initial traversal road, if a segment satisfies that the speed difference between the segment and the initial segment is less than a preset speed difference, designating the segment as a connected segment; if such segment that the speed difference between the segment and the initial segment is less than the preset speed difference is not found, acquiring a new initial segment from the segment sequence corresponding to the initial traversal road again, and searching for a new connected segment again according to the new initial segment;

designating a road corresponding to the connected segment as a new initial traversal road, and searching for a new connected segment again; and

repeatedly executing the above steps, to obtain the condition segments meeting the third preset condition from all the segment sequences.

According to the technical solution provided by the present disclosure, the acquiring a first road set of a planned route from a map navigation system of a vehicle, specifically includes the following steps:

acquiring an initial road set of the planned route from a map navigation system of a vehicle, wherein the initial road set at least includes a plurality of road segments and parameter data corresponding to each road segment identified by a global positioning system (GPS), and the parameter data at least includes a length;

traversing the initial road set, and labeling roads with lengths less than a first threshold as first-type roads; labeling roads with lengths greater than or equal to the first threshold as second-type roads;

normalizing all the first-type roads and roads connected with the first-type roads, calculating a first Euclidean distance and a second Euclidean distance between the normalized first-type roads and roads connected with the first-type roads, and selecting roads corresponding to the smaller Euclidean distance in the first Euclidean distance or the second Euclidean distance for combination, to obtain a plurality of segments of combined roads; meanwhile, denoting non-combined first-type roads as third-type roads; and

generating the first road set according to the combined roads, the third-type roads and the second-type roads.

According to the technical solution provided by the present disclosure, the first preset condition is as follows:

a difference between a segment orientation angle of the road segment and an orientation angle of a road corresponding to the road segment is less than a first difference, a difference between a segment length of the road segment and a length of the road corresponding to the road segment is less than a second difference, and a difference between a mean vehicle speed of the road segment and a mean vehicle speed of the road corresponding to the road segment is less than a third difference.

According to the technical solution provided by the present disclosure, the segment feature data at least includes: a mean speed of the segment, a mean acceleration of the segment, a mean deceleration of the segment and a confidence level of a speed-acceleration distribution. The second preset condition is as follows:

A difference between the mean speed of the road segment and a mean speed of the road corresponding to the road segment is less than a fourth difference, a difference between the mean acceleration of the road segment and a mean acceleration of the road corresponding to the road segment is less than a fifth difference, a difference between the mean deceleration of the road segment and a mean deceleration of the road corresponding to the road segment is less than a sixth difference, and the confidence level of the road segment is greater than a preset confidence level.

According to the technical solution provided by the present disclosure, the confidence level of the speed-acceleration distribution is acquired according to the following steps:

calculating a chi-square value of the speed-acceleration distribution;

acquiring a corresponding chi-square distribution database according to a freedom degree of the speed-acceleration distribution; wherein the chi-square distribution database at least includes a plurality of chi-square values and corresponding confidence levels; and

searching the confidence level corresponding to data associated with the chi-square values in the chi-square distribution database, which serves as the confidence level of the speed-acceleration distribution.

In a second aspect, the present disclosure provides a system for constructing a predictive vehicle driving condition, which can achieve the above method for constructing the predictive vehicle driving condition, and includes:

a data acquisition module, configured to acquire a first road set of a planned route from a map navigation system of a vehicle and retrieving historical vehicle driving data stored in the map navigation system; wherein the first road set at least includes a plurality of road segments identified by a global positioning system (GPS) and corresponding parameter data; and the historical vehicle driving data at least includes a plurality of point-in-time and corresponding time;

a data processing module, configured to calculate first distances from point-in-time of the historical vehicle driving data stored in the map navigation system to all roads in the first road set, identified by a global positioning system (GPS) to obtain a first distance set, and meanwhile, screen distances less than or equal to a distance threshold from the first distance set, to obtain a second distance set; and generate a road-point-in-time set according to the second distance set identified by the GPS; wherein the road-point-in-time set at least includes a plurality of road segments and a plurality of point-in-time corresponding to each road segment;

the data processing module, further configured to sequence a plurality of point-in-time corresponding to the same road segment in the road-point-in-time set according to a time sequence, to obtain a data sequence, meanwhile, traverse the data sequence, and cut at two adjacent point-in-time with a time difference greater than a time threshold, to obtain a plurality of road segments corresponding to a current road;

the data processing module, further configured to establish a road segment database of the road segment according to the segments meeting a first preset condition in the same road segment; wherein the road segment database at least includes a plurality of road segments and corresponding segment feature data;

the data processing module, further configured to screen segments with the segment feature data meeting a second preset condition from all the road segment databases, to generate a road-condition segment set; wherein the road-condition segment set at least includes a plurality of road segments and a plurality of condition segments corresponding to each road segment; and

the data processing module, further configured to sequence a plurality of condition segments corresponding to each road segment in the road-condition segment set according to a sequence from a high confidence level to a low confidence level, to obtain a plurality of segment sequences, and meanwhile, screen condition segments meeting a third preset condition from all the segment sequences, to generate an optimal driving condition; and outputting the optimal driving condition to a display and applying the optimal driving condition to drive the vehicle.

In a third aspect, the present disclosure further provides an electronic device, including a memory, a processor and computer programs stored on the memory and capable of running on the processor. When implementing the computer programs, the processor implements the steps of the above method for constructing the predictive vehicle driving condition.

In a fourth aspect, the present disclosure provides a computer readable storage medium on which computer programs are stored, and when the computer programs are executed by a processor, the steps of the above method for constructing the predictive vehicle driving condition are implemented.

In conclusion, the present disclosure discloses a specific flow of the method for constructing the predictive vehicle driving condition. According to the present disclosure, the distances from the longitudes and latitudes of the point-in-time of the historical vehicle driving data stored in the map navigation system to all roads in the first road set are calculated based on the acquired first road set of the planned route from a map navigation system of a vehicle and retrieving historical vehicle driving data stored in the map navigation system to obtain the first distance set, and meanwhile, the distances less than or equal to the distance threshold are screened from the first distance set, to obtain the second distance set; the road-point-in-time set is generated according to the second distance set identified by the GPS; the plurality of point-in-time corresponding to the same road segment in the road-point-in-time set are sequenced according to the time sequence, to obtain the data sequence; meanwhile, the data sequence is traversed, cutting is performed at two adjacent point-in-time with the time difference greater than the time threshold, to obtain the plurality of road segments corresponding to the current road; the road segment database of the road segment is established according to the segments meeting the first preset condition in the same road segment; the segments with the segment feature data meeting the second preset condition are screened from all the road segment databases, to generate the road-condition segment set; and the plurality of condition segments corresponding to each road segment in the road-condition segment set are sequenced according to the sequence from the high confidence level to the low confidence level, to obtain the plurality of road segments, and meanwhile the condition segments meeting the third preset condition are screened from all the segment sequences, to generate the optimal driving condition; and outputting the optimal driving condition to a display and applying the optimal driving condition to drive the vehicle.

According to the present disclosure, the first road set of the planned route from a map navigation system of a vehicle and retrieving the historical vehicle driving data stored in the map navigation system are combined, such that richer reference information is provided for constructing the vehicle driving condition, moreover, during the process of constructing the optimal driving condition, the road segment database corresponding to each road segment is finally generated by combining some roads, and cutting and screening the road segments. This results in more streamlined data in the road segment databases, which significantly reduces the calculation load and error rate during the construction process, and ensures that the final optimal driving condition is more accurate, thus achieving the objective of improving predictive accuracy. Consequently, the improvement in the vehicle driving range prediction algorithm and the predictive vehicle energy management strategy are effectively supported.

Reference signs in the drawings:: data acquisition module;: data processing module;

: electronic device;: central processing unit;: ROM (read only memory);: RAM (: read only memory;: random access memory);: bus;: I/O interface;: input part;: output part;: storage part;: communication part;: driver; and: removable medium.

The present disclosure will be further described below with reference to drawings and embodiments. It should be understood that the specific embodiments described herein are merely used for explaining relevant disclosures, instead of limiting the present disclosure. In addition, it should be noted that merely parts related to the present disclosure are shown in the drawings in order to facilitate the description.

It should be noted that the embodiments in the present disclosure and features in the embodiments may be mutually combined without conflicts. The present disclosure will be described in detail below with reference to the drawings and in conjunction with the embodiments.

is a flowchart of a method for constructing a predictive vehicle driving condition. Referring to, the method includes the following steps: